PROJECTS
Karner Blue Butterfly Predictive Habitat Modeling
The Karner blue butterfly is a federally endangered species, which only currently exists in a few states, including Wisconsin. Wisconsin Department of Natural Resources maintains a Habitat Conservation Plan (HCP) for the species, which includes a High Potential Range (HPR) map. I was asked to use my species distribution modeling (SDM) skills to develop a new HPR map in 2018. The final product was complete by spring of 2019 and went into regulation. The final shapefile for the HPR can be downloaded here.
LANDFIRE in the Northeast-Midwest
As the Northeast Region LANDFIRE Coordinator and research scientist in fire ecology at the Wisconsin Department of Natural Resources, I work on a wide variety of projects. You can see descriptions, status, and updates on some of those on the Google Forum I created.
One LANDFIRE project I will highlight an exciting collaboration between USFS, The Nature Conservancy, Michigan Technological University, Michigan Natural Features Inventory, and the Wisconsin Department of Natural Resources, working together on a fuels assessment in the Hiawatha National Forest (HNF). The project began as a vegetation departure analysis based on LANDFIRE data, and a large data gap was identified in the peatlands, which are abundant in the HNF. Peatland fires are arguably the most expensive to manage, and are very poorly understood. To address this need, my lab at the WI DNR is studying fire history in the forest through dendrochronology, and researchers at Michigan Technological University are sampling the peat for charcoal. Read more about it on the forum and in this press release.
My main duty as the regional LANDFIRE Coordinator is to gather the local expert knowledge I need to share with the LANDFIRE program. In order to meet these experts and expand my network, I present at many meetings each year. Some recordings are available below.
North Atlantic Fire Science Exchange's Igniting Exchange Conference (Portland, ME)
Video from a workshop I set up at the Stewardship Network Conference in Lansing, Michigan (resources and other videos available here).
For a bit more about some of the work I do, see my interview with the LANDFIRE program from my first year in the position.
Presenting at the SETAC conference
(Orlando, 2016)
Environmental Risk Assessment and related work
at Waterborne Environmental
Endangered species
The U.S. EPA and industry have been working for years to determine how to implement Section 7 of the Endangered Species Act as part of the pesticide registration process. I worked with a team to develop methodology for national scale risk assessments for endangered species.
Habitat modeling
One of the main issues in endangered species risk assessment is the lack of spatial habitat data. I led and managed multiple projects for modeling endangered species habitat at Waterborne Environmental. We tested species distribution modeling with MaxEnt and Random Forests (inductive modeling), and deductive modeling, where we set the tolerance of each species to environmental factors and overlay these datasets to find suitable environments. I used the Python and R programming languages to create these models.
Modelling Runoff of Pesticides in Soybean Production Areas in Brazil
The objective of this phase of the project was to estimate potential runoff mass fluxes and surface water concentrations for three chemicals that are used on soybeans in Brazil: lambda cyhalothrin, glyphosate, and three variations of thiamethoxam (A, B, and C). Model input scenarios were developed for all soil/weather/state combinations of soybean producing regions. To estimate environmental concentrations of pesticide residues in surface water, models were required that simulate runoff from fields into nearby ponds. The USEPA’s Office of Pesticide Programs’ (OPP) Pesticide Root Zone Model (PRZM) and their Exposure Analysis Modeling System (EXAMS) were used to estimate pesticide exposure for aquatic ecological concentrations.
Example results from the iSTREEM model
SETAC EU 2016
iSTREEM poster won first place at 2015 VAMLIS conference
Down-the-drain chemical modeling
Down-the-drain chemicals present their own unique issues in environmental impact modeling. Waterborne hosts the iSTREEM model for this purpose and I was heavily involved in making updates to the new version, including gathering and formatting public water system and stream location and attribute information for the spatial model. I was also a key member of the team estimating sewer residence times (the time it takes sewage to flow to a treatment plant) to spearhead accounting for biodegradation in the sewer as part of risk assessment for down-the-drain chemicals and solids, and presented our poster at the US SETAC conference in November 2016. See our publication in Science of The Total Environment here.
Web applications
I believe in the importance of empowering clients with interactive mapping products to assess impacts of potential regulations. I taught myself how to use ArcGIS Online and Web AppBuilder to address this need. See example here.
Cartography
I create many maps to display data, including this map to show Waterborne Environmental's global geospatial analysis work.
Open-source GIS
Most of my GIS work has been competed using Esri's ArcGIS software. This software is expensive and I am looking for ways to share data and empower people with spatial analysis who do not have access to ArcGIS. QGIS is a very powerful, free, open source GIS. I am learning how to use it so I can teach others. I am also learning how to share QGIS maps online so I can work with people that way. See my first map here, where I plot occurrence data for some species of interest to the Arlington Regional Master Naturalists.
Species distribution modeling
For my Master's project at the University of Virginia, I investigated the past range dynamics of a weed endemic to Europe that has become invasive in the United States. I performed population genetic analysis as well as species distribution modeling (SDM) to determine the species movements since the last glacial maximum (LGM) ~20,000 years ago. I presented parts of this project at several conferences, as pictured below.
Winning first place for student poster presentation at the Virginia Association for Mapping and Land Information Systems (VAMLIS) Conference
(Williamsburg, 2013)
Presenting at the Ecological Society of America Conference
(Minnesota, August 2013)
Presenting at the Esri User Conference
(San Diego, 2014)
Prescribed fire impact analysis
The Nature Conservancy uses prescribed burning in forests as a restoration technique. In the southern Appalachians, the forest attributes they are hoping to create with burning are 1) a widespread, open canopy and 2) patches of early successional forest. These attributes have not been common in recent prescribed fires. We wanted to understand how different environmental factors affect the outcome of the burns in order to plan for specific outcomes in future fires. We used aerial imagery to delineate canopy gaps in previous fires, and performed a regression with environmental factors obtained through GIS to understand the effect of each factor on fire severity.
Impacts of climate change in Mozambique
In my advanced spatial analysis class at UVa in the School of Architecture, I participated in a project to assess impacts of climate change in Mozambique in terms of climate scenario predictions, crop suitability, and flood modeling. I was chosen by the professor to continue the crop suitability modeling as a research assistant.
Predictive habitat modeling
As my senior research project at the University of Richmond, I created a predictive habitat model for the mountain monkey flower, Mimulus tilingii, using the Maximum Entropy software. Publication is in prep.
See an article about this project from the University of Richmond's spatial analysis lab.
Quantifying ecosystem services of a biofuel cropping system
As a biofuels research fellow through the Research Experience for Undergraduates (REU) at the Kellogg Biological Station (KBS), I conducted a project on mixed prairies as a potential biofuel. I assessed species richness, native species abundance, and flowering time in burned and unburned prairies in Michigan. I led a team of interns to record species observations in 1000m2 plots throughout the growing season.
